Reducing work interruptions and work-related interruptions of employees’ leisure time through job analysis and leadership coaching

Abstract

The study tested a brief intervention to stimulate and help supervisors reduce work-related interruptions among their employees, both at work and during leisure time. The core of the short-term intervention was a workplace analysis of work-related interruptions, which was fed back to supervisors in combination with a work redesign stimulation explaining why and how to reduce interruptions. Two intervention sessions, as one-on-one physical meetings, that lasted 1.5 h each and were 2 wk apart. The sample consisted of 20 managers and 89 employees. The non-experimental repeated measurement design comprised three questionnaire measurements of the 89 employees (two pre-measurements and one post-measurement). Repeated measure hierarchical linear models showed that the intervention significantly predicted reduced interruptions during work and work-related interruptions of leisure time. Although the intervention effect sizes were small, the current work design intervention with supervisors as mediating actors can reasonably contribute to occupational health prevention.

Introduction

Work interruptions can be described as events that hinder or delay achieving a goal at work¹⁾. A primary task is suspended temporarily because of the intrusion of a secondary, unplanned task. The new task is introduced by an event that can be either external or internal to the person, and there is an intention to return to and complete the primary task^{2, 3)}.

Interruptions are widespread in work settings. In Switzerland, about one-third of the employees report frequent or very frequent interruptions at work^{4, 5)}. In the European Union, 35% of all interruptions are considered disruptive⁶⁾. Frequent disruptive interruptions in the performance of one’s job are reported by 16% of all workers and 29% of managers in Europe⁶⁾.

Moreover, the frequency of work interruptions is increasing. For example, in Germany work interruptions almost doubled from 1998 to 2006⁷⁾.

Although work interruptions can be initiated externally and internally, this study focuses on externally triggered interruptions. There are many external sources of work interruptions: Coworkers drop by to seek help, share news, or socialize, and supervisors review work or assign new tasks. In addition, emails and other text or instant messages and phone calls from customers, clients, or other corporation partners can interrupt. Furthermore, equipment failures, such as computer breakdowns or lack of needed information, as well as distracting stimuli in the workplace, such as nearby conversations of others, can lead to disruptions of one’s behavioral performance, attentional focus, and concentration on an ongoing work task⁸⁾.

A sizeable representative survey in Germany found that 60% of employees with frequent interruptions perceived them as stressful⁹⁾. Thus, more than one in four employees (27%) is affected by work interruptions that induce stress. Accordingly, interruptions have become one of the most common stress factors in working life⁷⁾. Even if interruptions are not perceived as disruptive, resuming work after an interruption requires additional effort. The additional effort required varies depending on the activity and type of interruption.

Although interruptions may have positive effects under certain conditions¹⁰⁾, most previous research suggests that interruptions harm performance, well-being, and health. For instance, researchers have shown that interruptions can lead to forgotten intentions, quantitative and qualitative performance deterioration, and time loss^{11, 12)}. Time pressure and mental demands have been found to mediate the relationship between workflow interruptions and job satisfaction, as well as irritation⁷⁾. In some work settings, interruptions have even been discovered to be risky due to an enhanced likelihood of errors and accidents^13,14,15). In cross-sectional studies, researchers have established positive relationships between work interruptions and depression, psychosomatic complaints, and exhaustion in different occupational groups^16,17,18).

In their theoretical framework on work-related interruptions, Baethge et al.¹⁹⁾ explain how the accumulation of interruptions may lead to more intense and diverse work stressors and work strain. Frequent interruptions may change time loss into time pressure, and mistakes may become failures. Then, increased emotional stress and compensatory efforts may improve short-term performance but also lead to overstrain, incomplete recovery from work demands, and decreased performance quality in the long term. A downward spiral may result and explain the association between work-related interruptions and health problems found in applied research.

Supervisors can play a crucial role in reducing work interruptions for their employees in several ways. Wegge et al.²⁰⁾ describe five pathways through which a supervisor’s behavior can influence the health of organizational members. These include person-focused action, system-oriented action, action to mitigate the effects of contextual factors, climate control and identity management, as well as modeling. Especially in system-oriented action (but to a lesser extent in all other pathways), supervisor-driven work redesign is identified as a key factor for employee health that is often underestimated or overlooked in both theoretical and practical works²⁰⁾.

Supervisor-driven work redesign can eliminate or reduce work interruptions for their employees in several ways. First, they can introduce innovative solutions or redesign working conditions to minimize interruptions or reduce their disruptive effects^21,22,23). For example, introducing home office days, “quiet time”, fixed telephone duty hours, “office hours”, software solutions and guidelines, and designing the physical workplace environment can enable undisrupted work time. Second, training and raising employees’ awareness of the disruptive effects of work interruptions can raise awareness of each other’s interruptions²⁴⁾. Third, it is critical for leaders to empower employees to deal with work interruptions individually because people vary in their resilience regarding work interruptions^{25, 26)}. For example, an open-door culture is not for everyone and should not be required in general. Fourth, managers themselves can try to interrupt their employees less often, for example, by distributing tasks and fully communicating their concerns at pre-established times. Moreover, supervisors could be mindful of who they put on the distribution list for electronic messages. Finally, the supervisors’ own behavior and successful handling of interruptions can also act as a model, as subordinates may think supervisors expect similar behaviors from them. Therefore, an important pathway that should lead to fewer interruptions is better work design, planning, organization, and communication by supervisors, along with clear agreements between supervisors and employees. This should allow for more uninterrupted working hours.

Due to digitalization, many tasks can be performed regardless of location and time of day. Remote working and increasing expectations of work flexibility on the part of employees and employers have encouraged a blurring of the boundaries between work and leisure. This new way of working has also made it more common to have work-related interruptions during leisure time²⁷⁾. A strong urge to respond quickly to ICT-supported messages—so-called telepressure—can lead to prioritizing instant responses to messages and neglecting recovery time²⁸⁾. Telepressure has been shown to be a significant predictor of fatigue, reduced well-being, and worse overall health. In turn, it can contribute to predicting worse occupational health and well-being beyond task-related stressors and available resources²⁹⁾. The increasing use of technologies fostering 24/7 availability intensifies conflicts between work and privacy and impedes recovery from work^{30, 31)}. Work-related interruptions in leisure time make detachment from work more difficult^32,33,34,35). As detachment helps to recover from the demands of work³⁶⁾, recovery deteriorates as a result, which can negatively impact well-being and health. Studies indicated average positive correlations between detachment and self-reported mental (i.e., less exhaustion, higher life satisfaction, more well-being, better sleep) and physical (i.e., lower physical discomfort) health, state well-being (i.e., less fatigue, higher positive affect, more intensive state of recovery), and task performance³⁷⁾.

These work-related interruptions during leisure time by electronic mail or phone calls should also be reduced when task planning, organization, and communication are improved. Moreover, supervisors may also reduce sending (electronic) messages during office hours.

Interventions to reduce interruptions have been evaluated primarily in the health sector, but many evaluation studies have significant methodological shortcomings³⁸⁾. The vast majority of studies are based on a single-group design with two measurement points and thus offer only weak evidence regarding the effectiveness of the interventions. Consequently, there is a need for further development and evaluation of interventions to reduce interruptions, especially in office work, where interruptions are widespread³⁹⁾.

Therefore, the intervention examined in this research aims to sensitize, motivate, and enable supervisors to reduce work interruptions and work-related interruptions in leisure time. The intervention should only take a short time for the busy managers. Supervisors’ low effort and consultative approach should increase executive acceptance and compliance. Working directly with line managers also has the advantage of ensuring their commitment. Supervisors are often the driving forces behind change, and they make a decisive contribution to the success or failure of an intervention^{40, 41)}.

Through a mix of leadership coaching and training tailored to the individual situation of the respective supervisor and his or her team, the intervention should be specifically targeted and result in reducing interruptions. This reduction should be reached through better planning and organization of tasks and the workplace situation and through better communication.

The hypotheses of the study were therefore:H1: The intervention with supervisors reduces work interruptions in their employees.H2: The intervention with supervisors reduces work-related ICT-based interruptions of leisure time in their employees.

Subjects and Methods

Procedures and design

The intervention was a short-term leadership coaching/training, and the intervention group was the supervisor’s direct subordinates. The pre- and post-measurements were questionnaire surveys. A non-experimental design with two pre-measurements and one post-measurement was chosen. In contrast to a single-pre-measurement design, the two-premeasurement design could have ruled out alternative explanations for found intervention effects through history (unplanned events between measurements in addition to the intervention⁴²⁾). In addition, the present design allowed for differentiating between the effects of the questionnaire survey and the leadership coaching. This was important because each questionnaire survey should also be considered as an intervention with a potential effect⁴³⁾.

Three timeframes for the questionnaire surveys (pre-measurement 1, pre-measurement 2, and post-measurement) were established, within which the managers and their employees completed the online questionnaires. These were three time windows of several days, four weeks apart (pre-measurement 1 to pre-measurement 2) and eight weeks apart (pre-measurement 2 to post-measurement).

The participants were recruited by sending out informational materials by email. One hundred seven human resources managers from small and medium-sized enterprises (SME) and the federal and cantonal administration in the Bern area were contacted. Participation in the study was voluntary and was deemed harmless by the Ethics Committee of the Faculty of Human Sciences of the University of Bern (Application No. 2019-01-00005).

Intervention description

The coaches/trainers were industrial and organizational psychologists with expertise in results-oriented coaching methods⁴⁴⁾. The intervention aimed for supervisors to develop their own ideas for reducing interruptions. To do this, superiors need information about the current situation in their team from work analysis and knowledge about work design, which is based on action regulation theory⁴⁵⁾.

Supervisors were randomly assigned to two coaches. Two coaching/training sessions (1.5 h each) per supervisor were held at their workplace at two-week intervals. The sessions contain elements of training and coaching. Like typical training programs, they are based on achieving specific learning objectives. However, like coaching, they are oriented to an individual’s needs, characteristics, and experiences in a one-on-one setting.

The first session started 2 wk after the pre-measurement. Only supervisors attended the coaching sessions. In the run-up, the supervisors and their direct subordinates were surveyed online. The results on employee work interruptions and work-related interruptions during leisure time were discussed with the supervisors in the first coaching session. For this purpose, the coach analyzed, together with the managers, their own level of work interruptions and work-related interruptions during leisure time with the median values of their directly reporting team. Moreover, their direct reporting team’s median values were reported as percentile ranks in the distribution of a representative sample of Swiss workers in terms of age, gender, region, and industry. The benchmark comparison provided a frame of reference that helped to discuss the current level of work interruptions. Concerning work-related interruptions during leisure time, employees also reported whether they expected to be available for work problems even in their leisure time. In discussing work-related interruptions during leisure time, many supervisors reported that they had been unaware that their employees assumed they were expected to be available for work problems during leisure time. Hence, discussing the interruption data improved a shared mental model between supervisors and followers.

Together with the scores on the interruptions scale, scientific evidence regarding the adverse effects of interruptions was reported to increase supervisors’ motivation to address the issue. In addition, the concrete sources of interruptions were analyzed, ideas for reducing interruptions were collected, and advice on appropriate work and organizational design was given verbally and through compact training documents with concrete design tips.

At the end of each coaching session, the managers completed reflection sheets. The purpose was to stimulate reflection on the insights gained in the session and to develop and write down transfer goals and measures to achieve them. Before the second coaching session and at the end of the intervention, managers were emailed reflection sheets to remind them of their goals and review their progress toward achieving them. On the reflection sheets, we also asked about the motivation to pursue the transfer goals and requested feedback on the intervention.

Sample

The final sample of subjects who completed the questionnaires in full consisted of 20 managers and 89 employees. All participants were from the service sector and performed mainly office work. Of the 20 managers, 15 were male (75%) and 5 were female (25%), with ages ranging from 30 to 61 yr (M=44.85, SD=8.62). They had held a management position for between 0.5 and 22 yr (M=11.88, SD=7.71) and directly managed between four and 15 employees (M=7.65, SD=3.28) at the time of the study. Sixteen managers were employed full-time (one full-time work equivalent [FTE] in Switzerland corresponds to 42 h per wk). The remaining four were employed at least 80% (0.8 FTE). The highest level of education was attained by 15 managers at a university, two with a vocational apprenticeship and three with other training. Among the employees, 46 men (51.7%) and 43 women (48.3%) between the ages of 18 and 64 (M=42.52, SD=12.01) participated. Full-time employees accounted for 55.1%, 30.4% of the employees were employed between 80% and 90% (0.8–0.9 FTE), and the remaining 17.6% of the employees were employed between 20% and 70% (0.2–0.7 FTE). The highest level of education (university) was reported by 46.6% of the employees; 25.8% had an apprenticeship, 13.5% a vocational degree and 14.6% some other training.

Measures

Work interruptions as external unplanned events that hinder or delay achieving a goal at work^1,2,3) were assessed using a scale comprising three items from the Instrument for Stress-related Task Analysis (ISTA, Version 5.1)^46,47); e. g. “How often does it happen that you cannot work on something in a steady manner because something keeps getting in the way?”). Responses were given using a five-point Likert scale ranging from 1 (very rarely/never) to 5 (very often/constantly). Internal consistency (Cronbach’s α) was 0.688.

Work-related IT-based interruptions during leisure time were measured with a single item developed by Igic et al.⁴⁸⁾: “On average, how many times a day do you use your cell phone, tablet, or computer for work in your leisure time?”.

Control variables. Age, gender (female/male), and employment status (full-time vs. part-time) were included as control variables.

Motivation for working on transfer goals was asked on the reflection sheets with the question: “On a scale from 1 (not important) to 10 (very important), how important is it to you to achieve this transfer goal?

Intervention feedback was asked on the reflection sheet with the question: “How satisfied are you overall with the intervention (questionnaire survey and coaching session)?” to which responses were given using a series of seven faces showing expressions with written labels, ranging from a deep frown (1=very unsatisfied) to a giant smile (7=very satisfied)⁴⁹⁾.

Statistical analysis

The present study’s design yielded dependent data (three repeated measurements in the same sample), which had a hierarchical structure (individuals in teams). Hierarchical linear models with repeated measures in MLwiN 3.04 were used to meet these requirements in testing the hypotheses. Adjusting for age, gender, and full-time vs. part-time employment, significant differences in interruptions were expected in the employees between the pre- and post-measurements but not between the two pre-measurements. Age was grand-mean centered when it entered the multilevel-regression model that included three levels: Measurement at level 1, individuals at level 2, and teams at level 3. Note that unstandardized coefficients are reported. Reporting of effect sizes included standardized regression coefficients (beta coefficients). We used the restricted maximum-likelihood procedure to estimate the parameters. The alpha level was set to p<0.05, two-tailed.

Results

Descriptive results

The descriptive values for the 267 repeated measures (level 1), 89 individuals (level 2), and 20 different teams (level 3) can be found in Table 1.

Table 1. Descriptives for the three levels.

		Level 1 (measurement)		Level 2 (person)		Level 3 (team)
		M	SD	M	SD	M	SD
Age				42.52	12.01	42.44	6.69
Employment status				87.38	18.35	87.11	10.44
Work interruptions		3.3	0.7	3.3	0.65	3.32	0.4
Work interruption Preintervention 1				3.37	0.72	3.39	0.42
Work interruption Preintervention 2				3.31	0.71	3.33	0.38
Work interruption Postintervention				3.22	0.67	3.24	0.4
Leisure time interruptions		6.05	13.01	5.99	10.09	5.77	6.89
Leisure time interruptions Preintervention 1		7.35	14.69	7.35	14.69	7.15	8.09
Leisure time interruptions Preintervention 2		6.24	12.74	6.24	12.74	5.79	6.45
Leisure time interruptions Postintervention		4.59	11.42	4.59	11.42	4.39	6.04
Sex				n	%
	Female			43	48.3
	Male			46	51.7

Level 1: n=267, for leisure time interruptions n=263; level 2: n=89; level 3: n=20.

M: Mean; SD: standard deviation.

The Pearson correlations for work interruptions with work-based leisure time interruptions were weak and not significant on all levels (L1: r=0.017; L2: r=0.025; L3: r=0.013).

At Level 2 and Level 3, correlations with sex, age, and employment status were calculated. The only significant correlations we found were with sex. Sex correlated significantly (p<0.01) with work-related interruptions of leisure time, and this at medium strength at Level 2 (r=0.324) and strong at Level 3 (r=0.645): Male sex is related to more frequent work-related interruptions of leisure time. Furthermore, sex and full-time employment correlate significantly in the same direction (L2: r=0.305; L3: r=0.50): Male gender goes along with more frequent full-time employment. The (nonsignificant) correlations of sex with work interruptions, on the other hand, are small and negative (L2: r=−0.137; L3: r=−0.219). The correlations between sex and age are also small (L2: r=0.025; L3: r=−0.078). Age correlates weakly (negatively) with work interruptions on Level 2 (r=−0.052) and on Level 3 (r=−0.191 and with work-related interruptions of leisure time positively on Level 2 to r=0.137 and on Level 3 to r=0.102). The correlations of employment status with work interruptions on Level 2 were small and negative (r=−0.096) and on Level 3 of medium height (r=−0.307), and concerning work-related interruptions of leisure time positive but small (L2: r=0.086; L3: r=0.179).

At Level 1 (repeated measures), correlations with the intervention can also be examined (0=before the intervention; 1=after the intervention). The correlations of the intervention are small: r=−0.077 with work interruptions and r=−0.080 with work-related interruptions of leisure time.

Results for tests of H1 and H2 from multilevel-regression analyses

The multilevel regression models with 267 repeated measures (Level 1) for 89 individuals (Level 2) of 20 different teams (Level 3) showed a significant intervention effect on work interruptions (two-sided test; B=−0.101, SE=0.036, β=−0.07, p=0.006), controlling for age, gender, and full-time vs. part-time employment (Table 2). In line with the first hypothesis, the intervention reduces work interruptions in participants.

Table 2. Multilevel regression: Prediction of work interruptions and work-related interruptions of leisure time.

	Work interruptions						Leisure time interruptions
	B	SE	t	p	N	IGLS	B	SE	t	p	N	IGLS
constant	3.585	0.136	26.36	<0.001			3.810	1.679	2.269	0.024
Agea (yr)	−0.002	0.007	−0.286	0.775			0.126	0.087	1.448	0.149
Sex (female=0, male=1)	−0.100	0.160	−0.625	0.533			5.685	2.530	2.247	0.025
Intervention (0=no, 1=yes)	−0.101	0.036	−2.778	0.006			−2.253	0.939	−2.400	0.017
Part-time work (0=no, 1=yes)	−0.261	0.165	−1.582	0.115			0.156	1.916	0.082	0.935
Level 3 (Team)	0.047	0.053	0.887	0.376	20		3.731	5.964	0.626	0.532	20
Level 2 (Person)	0.473	0.079	5.987	<0.001	89		57.120	15.690	3.641	<0.001	89
Level 1 (Measurement)	0.079	0.008	9.875	<0.001	267	325.724	99.490	11.020	9.028	<0.001	263	2023.971

A full-time work equivalent in Switzerland corresponds to 42 h per wk. ICC (Intra-class correlation coefficient) = Percentage of variance between persons (ICC = variance between persons / [variance between persons + variance within]); in work interruptions ISS is 7.8% for L3, 79.4% for L2, and 12.8% for L1. In work-related interruptions of leisure time ICC is 0.52% for L3, 36.29% for L2, and 63.19% for L1. p=significance level (probability that B is zero), IGLS: iterative generalized least squares. The model is controlled for age, gender, and part-time work. SE: standard error. ^agrand-mean centered.

Moreover, work-related ICT-based interruptions during leisure time were predicted to be lower after the intervention than before (two-sided test; B=−2.253, SE=0.939, β=−0.08, p=0.017), controlling for age, gender, and employment percentage (Table 3). In line with the second hypothesis, the intervention reduced work-related ICT-based interruptions during leisure time in followers.

Table 3. Multilevel regression: prediction of work-related interruptions of leisure time.

	B	SE	t	N	IGLS
Age	0.126	0.087	1.448
Sex (female=0, male=1)	5.685	2.530	2.247
Intervention (0=no, 1=yes)	−2.253	0.939	−2.400
Constant	3.810	1.679	2.269
Part-time work (0=no, 1=yes)	0.156	1.916	0.082
Level 3 (Team)	3.731	5.964	0.626	20
Level 2 (Person)	57.120	15.690	3.641	89
Level 1 (Measurement)	99.490	11.020	9.028	263	2023.971

A full-time work equivalent in Switzerland corresponds to 42 h per wk. ICC (Intra-class correlation coefficient) = Percentage of variance between persons (ICC = variance between persons / [variance between persons + variance within]); ICC is 0.52% for L3, 36.29% for L2, and 63.19% for L1. IGLS: iterative generalized least squares. The model is controlled for age, gender, and employment status. SE: standard error.

Motivation to work on transfer goals and intervention feedback

After the second coaching session, when asked how important their transfer goal was to them, managers rated on average 9.00 (SD=0.94) on a scale of 1 (not important at all) to 10 (very important). After the end of the intervention, when asked how motivated they were to pursue their transfer goal since the last coaching session, the mean score was 8.28 (SD=1.41).

One month after completion of the intervention, 78% of the supervisors reported being very or exceptionally satisfied with the intervention. The remaining supervisors stated that they were pretty satisfied with the intervention.

Discussion

This study aimed to show that a brief intervention addressing supervisors can reduce interruptions among their subordinates at work and during leisure time. The intervention was shown to be significantly associated with both reduced work interruptions during work and reduced work-related ICT-based interruptions of leisure time among followers.

The intervention investigated was directed at supervisors but aimed to improve the working conditions of their subordinates, as it was assumed that supervisors influence their employees’ interruptions in various ways. Supervisors can design working conditions, and their careful planning and communication can prevent interruptions both during working hours and during leisure time. In addition, supervisors act as role models. Through workplace analysis and benchmark comparison, it was surveyed and shown to the supervisors how much their employees are affected by interruptions compared to a representative sample of employees in Switzerland. By illustrating that reducing interruptions seems to have a significantly positive impact on performance, well-being, and health, supervisors should be additionally motivated to take this issue seriously and strive actively to improve their employees’ working conditions and align their planning, organizational, and role-model behavior accordingly. The results of the intervention feedback questionnaires from supervisors, which were completed by the supervisors several times during the intervention, show that reported motivation to work on the transfer goals was indeed high. Working with line managers to ensure their engagement stands out as a benefit of this intervention study and is consistent with Nielsen⁴⁰⁾, who reports that process evaluations of organizational intervention studies indicate that line managers are critical contributors to the success or failure of an intervention as drivers of change.

Because we worked with supervisors rather than directly with their subordinates, we delivered the intervention in an individual, one-on-one setting with comparatively little effort. Among other things, the individual setting has the advantage that supervisors can talk openly about their difficulties, convictions, and areas of development in a trusting atmosphere and work very specifically and with a focus on individually relevant topics. Compared to interventions in group settings, this not only means less effort for the participants in terms of duration but should also further increase their motivation. The fact that there was not a single drop-out during the study can also be seen as a corresponding indicator of motivation and compliance and a strength of the study.

The intervention also reduced work-related interruptions during leisure time, which should facilitate employees’ boundary management. Work-related activities due to interruptions during leisure times not only disrupt leisure activities but also direct attention towards work. Thus, such interruptions hamper detachment from work during leisure time, which helps recover from work³⁷⁾ and is a goal of boundary management. Noteworthy, the study took place before the COVID-19 pandemic. With the COVID-19 pandemic, boundary management has become even more demanding, primarily because of the increase in working from home, where avoiding interference with work and private issues becomes more challenging⁵⁰⁾.

It, therefore, seems essential not only to leave it up to the employees themselves to get their work behavior as well in their leisure time under control. Superiors should also be a contributing and supporting force. Research so far has often assumed that boundary management is predominantly a matter of self-regulation. The present study, however, showed that the supervisor can also influence electronic media use for work during leisure time. Even if the supervisor has not explicitly communicated it, many employees think the supervisor expects employees to be available for work-related matters during their leisure time.

Moreover, the supervisor’s own handling of interruptions as a role model may already signal the legitimacy of dealing with work issues during leisure time. In addition, presumably, simply the fact that employees receive electronic messages from their supervisor outside of office hours makes them assume that he/she expects them to be accessible during leisure time. The intervention made many supervisors realize how often they themselves interrupt their employees in their leisure time (and at work) and what costs these interruptions may have. Finally, supervisors realized that interrupting their subordinates during leisure time reflects their own insufficient boundary management.

Hence, in the current intervention study, supervisors acted as an intermediary. Montano et al.⁵¹⁾ emphasize the need for evidence-based leadership interventions incorporating workplace health. Awareness of supervisors of problematic work conditions is often incomplete, as is awareness of the availability expectations of followers. Therefore, in the current study, employees were asked about their work situation, and the coach discussed job analysis results with supervisors. Therefore, the intervention itself targeted the supervisors and aimed to reduce the interruptions of the employees. The results confirm that it is possible to reduce both interruptions at work and work-related interruptions of employees’ leisure time through a relatively minimal intervention involving only two 1.5-h coaching/training sessions for their supervisors in a one-to-one setting.

The effect sizes were small. As emphasized by Tanner-Smith et al.⁵²⁾, general benchmarks for interpreting effect sizes as small/medium/large should be abandoned, as the magnitude of intervention effect sizes should be assessed relative to the context of the intervention area. Because we are not aware of sufficiently similar intervention studies, appropriate comparisons are difficult. Rigotti³⁸⁾ concludes, after reviewing the available intervention studies on reducing interruptions in work settings, that there is some evidence of successful interventions to reduce interruptions in medical and nursing activities that, taken together, suggest a real but relatively small effect. Meanwhile, small effects seem to be realistic in this research field. A meta-analysis by Robertson et al.⁵³⁾ reports that the effects of intervention studies on organizational changes are usually small to moderate. The benefits and success of interventions at the organizational level may depend on different workplace contexts⁵⁴⁾ and specific conditions such as sufficient and continuous management support, appropriate problem assessment so that the intervention fits the problems to be solved, and the active involvement of employees⁵⁵⁾. Contextual changes such as restructuring, downsizing, high turnover among managers or employees, competing projects, and similar incidents can also affect the chances of positive outcomes^{54, 55)}.

On the background of expectably small effects, the effort for the present type of intervention seems reasonable. Although small in effect size, the reduction of interruptions is meaningful in the participants’ experience. This is shown by the large proportion of managers who report being very or exceptionally satisfied with the intervention one month after completion of the intervention. Comparing the effort of the intervention and its effects, one can draw a positive balance. On the one hand, there is the measurement effort before the actual intervention, as all employees have to be asked about their work situation by questionnaires. While the actual intervention is limited to the supervisors, it may affect a large number of subordinates. Since it takes place in a one-to-one setting, the areas for development of the managers can be addressed in a targeted way, and thus, with a duration of 2 times 1.5 h, they can be effectively trained and coached according to their needs. The very short time frame of the intervention, which took place at the supervisor’s workplace so that no additional time was required for traveling, is very convenient and motivating for busy managers.

Ely et al.⁵⁶⁾ noted that most studies evaluating leadership coaching rely on leaders’ self-assessment, which is a major limitation in the literature evaluating leadership coaching. In contrast, the present study uses the assessment of relevant third parties (namely, the direct subordinates of the supervisor) to evaluate the effectiveness of changes in leadership behavior. Moreover, it is not the leadership behavior that is assessed, but rather the impact of changes in coachees’ leadership behaviors on subordinates’ working conditions.

The intervention also aims to improve working conditions. The research on prevention and interventions regarding occupational health conducted to date has targeted employees more often than work situations, as work-related measures are more complex and challenging to implement⁵⁷⁾. As the intervention is conducted per team or company unit, it can be applied regardless of the company’s overall size. Due to the individual component of the one-on-one setting with the supervisors, it can also be applied universally across all business sectors or industries, which is another advantage of this kind of intervention.

The ultimate, longer-term goal of the intervention is to improve performance, well-being, and health by reducing interruptions. Work interruptions can harm performance, well-being, and health^{7, 14, 16, 17)}. This also refers to work-related interruptions in leisure time. These can increase work-privacy conflicts, impair detachment from work, and thus make recovery from work more difficult^32,33,34,35), which can affect health, well-being, and task performance³⁷⁾. Reducing interruptions, both during working hours and during leisure time, should thus help promote employees’ well-being, health, and performance. Thus, future studies should record how much the reduction of interruptions improves performance, well-being, and health and reduces errors and accidents in the longer term.

Reducing interruptions may also have negative consequences because work interruptions may have positive consequences, depending on several factors, such as who interrupts or the content and function of the interruption¹⁾. For example, a supervisor who provides an employee with missing information in real-time or provides immediate informal feedback can avoid problems and wasted time in the future and help drive performance and relationships. This informal feedback and information exchange often does not occur through other channels and should, therefore, not be prevented but regarded as an interruption of work with positive consequences¹⁾.

The positive effects of the current intervention could be expected to be higher in very complex tasks. For example, the results of an experimental study show that interruptions facilitate performance on simple tasks while inhibiting performance on more complex tasks. However, even “helpful” interruptions that facilitate completing simple tasks are often perceived negatively⁵⁸⁾. Both characteristics of the interrupted primary task and the individual appraisal of work interruptions can be relevant to the subjective perception of work interruptions as an overload⁵⁹⁾. Accordingly, reducing work interruptions might not always be perceived as reducing interruptions overload (for instance, when a change towards more complex work tasks occurs simultaneously). Future research investigating the effects of a reduction in interruptions should consider both positive and negative consequences of interruptions, the role of the complexity of the interrupted tasks, and the subjective evaluation of work interruptions as overload.

In summary, the current intervention has the potential to reduce unnecessary and unreasonable interruptions while sound interruptions are kept unchanged. This expectation is based on the job analysis as the basis of the intervention that should result in work redesign that can reduce unnecessary and unreasonable interruptions²⁷⁾.

The sources, content, and function of interruptions were not distinguished more deeply in this study. Is it interruptions from supervisors, colleagues, or external parties such as customers and suppliers that are reduced by the intervention? Does the intervention also have an impact on self-interruptions, even though it does not directly affect them, since it is not the employers themselves who are being coached, but their supervisors? It would be interesting to distinguish between source contents and functions of the interruptions, to get a more differentiated picture of which interruptions are reduced by this kind of intervention and which further interventions might additionally promise success.

Furthermore, it is well conceivable to expand the intervention approach and focus not only on work interruptions but also on other work stressors and resources. Detailed knowledge about the sources and nature of work interruptions and other stressors and available resources could help managers derive more targeted work design arrangements.

Finally, further intervention studies to improve working conditions and reviews comparing the effects of interventions aimed at supervisors with those aimed directly at employees seem desirable.

Limitations

Our study is not without its limitations. Bias from questionnaire responses as common source variance may have booted correlations in this study⁶⁰⁾. In addition, the sample size of 20 supervisors and 89 followers is small. In addition, sample selection is a general problem in social research that occurs when a random sample is not drawn, as in the present study. It is a threat to making valid causal inferences in intervention research because, for example, individuals with more significant problems are more likely to be selected for intervention⁶¹⁾. However, the fact that when recruiting for participation in the study, a reduction in interruptions was not explicitly mentioned, but only an improvement in working conditions could have counteracted this problem somewhat.

Work-related interruptions in leisure time were captured by a single item. Major criticisms of single-item measures include lack of content validity due to criterion deficiencies and unreliability^62,63,64). However, based on three studies, Fisher et al.⁶⁵⁾ showed that although multiple-item measures are preferable from a psychometric perspective, single-item measures can also provide helpful information in organizational research. Furthermore, with the measurement of work-related interruptions in leisure time with an item asking for a daily number of times using electronic devices in the leisure time for work, it is not clear whether the interruption of free time for work is self-determined or externally imposed. Research on work-related interruptions in leisure time also often fails to distinguish whether the interruption is self-induced (for example, checking emails on Sundays at discretion) or induced by an external event (for example, a call from a boss while bowling with friends). One exception is the study by Khalid et al.⁶⁶⁾, which showed that interruptions initiated by others intensify the relationship between after-hours work-related technology use and its outcomes such as work-family conflict and deviance. The authors explain the more detrimental effects of externally initiated interruptions by the fact that these types of interruptions deplete available free resources and create a situation in which the perception of loss of control triggers harmful reactions⁶⁶⁾. Either way, with the increasing use of electronic devices for work in leisure time, the boundaries between work and leisure become more permeable, and mental detachment is hindered. It would be helpful to demonstrate the relationship between the number of (self-initiated and others-initiated) occupational uses of electronic devices in leisure time and well-being and health in prospective studies.

Finally, one limitation of our study is that we did not ask comprehensively which actions superiors had really implemented to reduce interruptions. This should be done more systematically and in detail in subsequent studies. After the intervention, some supervisors reported that they became more aware of their role model function and subsequently changed their behavior; for example, they closed their doors more often. Others reported paying more attention to sending emails during office hours. Others reported that they addressed reducing interruptions in team meetings or short workshops and introduced subsequent measures such as rotating telephone service hours.

Conclusion

The expectation that employees’ work interruptions and work-related interruptions in their leisure time can be reduced by providing appropriate feedback to supervisors was confirmed.

Although the effects are small, this should be put in relation to the high availability of the intervention due to the relatively low effort required for the intervention.

To date, work design has received only marginal attention in leadership research. The intervention examined in this study, which combines job analysis and the benefits of leadership coaching and training, offers an excellent opportunity to link job analysis and design with leadership development.

Since work interruptions are among the most common workplace stressors and are related to well-being and health, this intervention effectively contributes to workplace health promotion.

Conflict of Interest

No potential conflict of interest was reported by the authors.